{"id":"W4405746722","doi":"10.1016/j.neunet.2024.107067","title":"Promises and perils of using Transformer-based models for SE research","year":2024,"lang":"en","type":"review","venue":"Neural Networks","topic":"Software Engineering Research","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Fundamental Research Funds for the Central Universities; Sun Yat-sen University","keywords":"Transformer; Computer science; Artificial intelligence; Machine learning; Engineering; Electrical engineering; Voltage","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001006763,0.0003086679,0.0008850517,0.0004419102,0.0001036395,0.0002498014,0.0009269046,0.0002839481,0.00000172801],"category_scores_gemma":[0.0001268426,0.0002387224,0.0003055721,0.00109122,0.0001228451,0.000225613,0.0001691714,0.000895507,8.508059e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008936795,"about_ca_system_score_gemma":0.0003512323,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001248409,"about_ca_topic_score_gemma":0.000001500337,"domain_scores_codex":[0.9975908,0.0001621847,0.0004324512,0.000654986,0.0005008222,0.0006587959],"domain_scores_gemma":[0.9969313,0.002144331,0.00005765984,0.0005439633,0.0001758618,0.0001468924],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000006168138,0.00001585734,7.078094e-7,0.02154019,0.00004318889,0.00001430463,0.00005829617,0.03525437,8.733023e-7,0.0004074605,0.0003017028,0.9423569],"study_design_scores_gemma":[0.0001044151,0.0001093667,1.963007e-7,0.005336471,0.0000784617,0.00002186463,0.000001087164,0.938386,0.000002265049,0.0001944589,0.05556744,0.0001980208],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.00002021086,0.7086973,0.2897369,0.00002224867,0.00024221,0.001160584,0.00001312661,0.00009941675,0.000008023186],"genre_scores_gemma":[0.001010197,0.9888102,0.009466394,0.000006403539,0.0002571728,0.0002942492,0.00001143466,0.00008392787,0.00005998371],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.9421589,"threshold_uncertainty_score":0.9734819,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3153174685816854,"score_gpt":0.4532480130661218,"score_spread":0.1379305444844364,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}